The present revised application seeks 3 years of support to test the efficacy of a promising psychosocial smoking prevention strategy both with and without additional school-wide support activities. A cohort of urban and suburban 7th graders (N=6,000) from 18 schools would be asigned to the following conditions: (1) smoking prevention curriculum alone, (2) smoking prevention curriculum plus school-wide support activities, and (3) control. The schools assigned to conditions 1 and 2 would receive an 18 session smoking prevention program in grade 7 and a 10 session "booster" curriculum in grades 8 and 9. The proposed study would extend our previous research by examining the efficacy of the LST prevention strategy with urban students from minority and lower SES families, by testing the extent to which interventions targeting at the larger school environment can help maximize the effectiveness of the prevention curriculum, and by providing follow-up throughout the critical junior high school period. All students will be pre- and post-tested during the first project year with additional follow-up post-tests occurring semi-annually thereafter. Students will be assessed by questionnaire for self-reported smoking status and behavioral intention to smoke as well as for several cognitive, attitudinal, and personality predisposing variables. Saliva samples will be collected and analyzed in a 20% subsample to determine SCN levels. Chi-square analyses will be used to analyze smoking status data, and ANCOVAs will be used to examine changes between conditions in terms of the predisposing variables. Discriminant function analyses and causal modeling techniques will be utilized to examine the adequacy of the theoretical model guiding this research.